1. The PAR-6 Polarity Protein Regulates Dendritic Spine Morphogenesis through p190 RhoGAP and the Rho GTPase 
Huaye Zhang, Ian G. Macara 
Developmental Cell 
Volume 14, Issue 2, Pages (February 2008)
DOI: /j.devcel
Copyright © 2008 Elsevier Inc. Terms and Conditions

2. Figure 1 PAR-6C Is Necessary for Spine Maintenance
(A) PAR-6C localizes to excitatory synapses. Hippocampal neurons were fixed and coimmunostained with PAR-6C and PSD-95 antibodies.
(B) Efficiencies of PAR-6C shRNA constructs. Rat2 fibroblasts were electroporated with pSUPER-luciferase (Control), pSUPER-PAR-6C shRNA#3, or pSUPER-shRNA#8 (PAR-6C KD). 48 hr after transfection, cells were lysed and analyzed by SDS-PAGE and western blotting.
(C) Effects of PAR-6C knockdown on spine formation. Hippocampal neurons were transfected with either pSUPER-luciferase (Control) or pSUPER-PAR-6C shRNA#3 (PAR-6C KD) at DIV 6–7 and imaged at DIV 14. Yellow fluorescent protein (YFP) was coexpressed as a marker for transfected neurons and to visualize cell morphology. Knockdown of PAR-6C significantly reduced the number of spines. ∗p < by Student's t test.
(D) PAR-6C is necessary for spine maintenance. Hippocampal neurons were transfected with either pSUPER-luciferase (Control) or pSUPER-PAR-6C shRNA#3 (PAR-6C KD) at DIV 17 and imaged at DIV 22. Yellow fluorescent protein (YFP) was coexpressed as a marker for transfected neurons and to visualize cell morphology. There was a significant reduction in the number of dendritic spines even though PAR-6C was knocked down after the neurons had formed numerous spines, suggesting that PAR-6C is necessary for spine maintenance. ∗p < by Student's t test.
Error bars represent SEM of three independent experiments.
Developmental Cell  , DOI: ( /j.devcel )
Copyright © 2008 Elsevier Inc. Terms and Conditions

3. Figure 2 Overexpression of PAR-6C Promotes Spine Morphogenesis
(A) Effects of PAR-6C overexpression on spine formation. Hippocampal neurons were transfected with either an empty myc vector or myc-tagged PAR-6C. YFP was coexpressed to visualize the transfected cells. Overexpression of PAR-6C increases the density and length of spines. ∗p < by Student's t test.
(B) Effects of PAR-6C on the number of functional synapses as visualized by FM1-43 dye uptake.
(C) Quantification of the results shown in (B). ∗p < by Student's t test.
Error bars represent SEM of three independent experiments.
Developmental Cell  , DOI: ( /j.devcel )
Copyright © 2008 Elsevier Inc. Terms and Conditions

4. Figure 3 Effects of PAR-6 Mutants on Spine Morphogenesis
(A) Schematic diagram of PAR-6 mutants used in this study.
(B) Effects of overexpressing different PAR-6 mutants on spine formation. Hippocampal neurons were transfected with the indicated constructs. YFP was coexpressed to visualize the transfected cells.
(C) Quantification of spine density in neurons expressing different PAR-6 mutants. ∗p < by Student's t test.
(D) The aPKC binding domain and the PDZ domain are necessary for rescuing the PAR-6C knockdown defect. Hippocampal neurons were transfected with pSUPER-luciferase (Control), pSUPER-PAR-6C shRNA#3, or shRNA #3 together with the indicated PAR-6 constructs. YFP was coexpressed to visualize the transfected cells.
(E) Quantification of the rescue results shown in (D). ∗p < by Student's t test.
Error bars represent SEM of three independent experiments.
Developmental Cell  , DOI: ( /j.devcel )
Copyright © 2008 Elsevier Inc. Terms and Conditions

5. Figure 4 PKCζ Functions Downstream of PAR-6 in Spine Morphogenesis
(A) Effects of different PKCζ mutants on spine morphogenesis. Hippocampal neurons were transfected with YFP and the indicated PKCζ mutants. A kinase-active PKCζ (PKCζ T410E) promotes spine formation, whereas a kinase-inactive mutant (PKCζ T410A) inhibits spine formation.
(B) Quantification of the results in (A). ∗p < by Student's t test.
(C) Hippocampal neurons were transfected with YFP and the indicated constructs. PKCζT410E cannot rescue the spine formation defects in PAR-6C knockdown neurons or the defects in neurons expressing PAR-6BmutPDZ, indicating that targeting information provided by PAR-6 is necessary for the function of aPKC.
(D) Quantification of the results in (C). ∗p < by Student's t test.
(E) Localization of PKCζT410E depends on the presence of PAR-6. The localization of PKCζT410E was examined in control neurons and neurons depleted of PAR-6. PKCζT410E localizes to spines in control neurons but fails to accumulate in spines in neurons depleted of PAR-6.
(F) PKCζ functions downstream of PAR-6. Hippocampal neurons were cotransfected with PAR-6B and either PKCζT410E or PKCζT410A. Coexpression of PAR-6 and PKCζT410E does not increase spine number to a level higher than each construct singly expressed. PKCζT410A efficiently inhibited spine formation induced by PAR-6B.
(G) Quantification of the results shown in (F). ∗p < by Student's t test.
Error bars represent SEM of three independent experiments.
Developmental Cell  , DOI: ( /j.devcel )
Copyright © 2008 Elsevier Inc. Terms and Conditions

6. Figure 5 RhoA Functions Downstream of PAR-6 in Spine Morphogenesis
(A) Effects of RhoA mutants on spine morphogenesis. Hippocampal neurons were transfected with YFP and the indicated RhoA constructs. Constitutively active RhoA (RhoA V14) inhibits spine formation, whereas dominant-negative RhoA (RhoA N19) promotes spine formation. ∗p < by Student's t test.
(B) Dominant-negative RhoA rescues the spine formation defects in PAR-6C knockdown neurons. Hippocampal neurons were transfected with YFP and the indicated constructs. RhoA N19 reverses the spine formation defects in PAR-6C shRNA#3-expressing neurons. ∗p < by Student's t test.
(C) Treatment of PAR-6C knockdown neurons with a ROCK inhibitor H-1152 (10 μM) reverses the spine formation defects in these neurons. ∗p < by Student's t test.
(D) Constitutively active RhoA inhibits spine formation induced by PAR-6 overexpression. Hippocampal neurons were transfected with YFP and the indicated constructs. RhoA V14 inhibited spine formation induced by PAR-6B. ∗p < by Student's t test.
(E) PAR-6 affects RhoA activity. Hippocampal neurons were transfected with an empty vector, myc-tagged PAR-6C, or pSUPER-PAR-6C shRNA#3. A pRaichu-RhoA FRET biosensor was coexpressed to examine RhoA activity in these neurons. Representative FRET images are shown.
(F) Quantification of mean FRET values shown in (D). ∗p < by Student's t test.
Error bars represent SEM of three independent experiments.
Developmental Cell  , DOI: ( /j.devcel )
Copyright © 2008 Elsevier Inc. Terms and Conditions

7. Figure 6
p190 RhoGAP Functions Downstream of PAR-6/aPKC in Spine Morphogenesis
(A) Coimmunoprecipitation of endogenous PKCζ and p190 RhoGAP from hippocampal neurons. Hippocampal neurons (DIV 8) were lysed and immunoprecipitated with either a GFP antibody or a PKCζ antibody. Immunoprecipitates were washed and analyzed with SDS-PAGE and western blot. IP: immunoprecipitation. WB: western blot.
(B) Efficiency of the p190A RhoGAP shRNA construct. Hippocampal neurons were transfected with YFP and pSUPER-p190A shRNA. At DIV 14 cells were fixed and immunostained for p190 RhoGAP. The staining intensity was significantly reduced in pSUPER-p190A shRNA expressing neurons as compared with nearby untransfected neurons. IF: immunofluorescence.
(C) Knockdown of p190 RhoGAP reduces spine density. Hippocampal neurons were transfected with either pSUPER-luciferase or pSUPER-p190 shRNA. YFP was coexpressed to visualize the transfected cells. ∗p < by Student's t test.
(D) The GAP activity of p190 RhoGAP is necessary for spine formation. Hippocampal neurons were transfected with an empty vector (Control), wild-type p190 RhoGAP, or a GAP-deficient mutant of p190 RhoGAP (p190 RhoGAP R1283A). YFP was coexpressed to visualize the transfected cells. ∗p < by Student's t test.
(E) P190 RhoGAP functions downstream of PAR-6. Hippocampal neurons were cotransfected with PAR-6B and either wild-type or GAP-deficient p190 RhoGAP. The GAP-deficient p190 RhoGAP inhibited the spine formation induced by PAR-6B overexpression. ∗p < by Student's t test.
(F) PAR-6 regulates RhoA activity through p190 RhoGAP. Hippocampal neurons were transfected with either an empty vector plus pSUPER-p190 shRNA, or myc-PAR-6C plus pSUPER-p190 shRNA. pRaichu-RhoA was coexpressed to visualize RhoA activity. Representative FRET images are shown.
(G) Quantification of mean FRET values shown in (F). ∗p < by Student's t test.
Error bars represent SEM of three independent experiments.
Developmental Cell  , DOI: ( /j.devcel )
Copyright © 2008 Elsevier Inc. Terms and Conditions